Little’s Law: find the work waiting in your business

Choose a boundary before doing the arithmetic

A repair shop might measure from a customer approving a repair to the repaired item being ready for collection. It should not count enquiries as open repairs while measuring completions only for approved jobs. Those are different populations.

The relationship is usually written L = λW. L is average work in progress, λ is the average flow rate and W is average elapsed time. The Illinois Business Innovation Services teaching material presents the same relationship for production flow.

Use consistent units and a representative period.

For a growing backlog, a short measurement window or a seasonal rush, the simple calculation needs care. It is an average relationship, not a promised completion date for every customer.

A repair queue in numbers

Consider an illustrative shop with an average of 24 approved repairs open and eight repairs completed per working week. Average elapsed time within that boundary is 24 ÷ 8 = three working weeks. A repair requiring two hours at the bench can still spend weeks waiting.

Suppose the shop reduces average open repairs to 16 while sustaining eight completions a week. The corresponding average becomes two weeks. Compare accepted work, margin and service quality before extending the change.

Customers may have left, job complexity may have changed or technicians may be working overtime.

Find what the queue is waiting for

Take the ten oldest jobs and ask one concrete question about each: what prevents the next step today? Useful reasons include awaiting a part, awaiting customer approval, awaiting a technician and awaiting a quality check. Give the reason an owner and a next review date.

The Lean Enterprise Institute’s standardized-work guidance includes process-capacity analysis. Here, the practical task is to separate a genuine shortage of skilled bench hours from missing instructions or parts. Adding a technician will not resolve an unapproved quotation.

Connect the queue to cash with care. Parts bought for unfinished jobs and unissued invoices can tie up money. Lead time alone does not calculate that amount.

Record deposits, purchased materials and billing milestones separately.

Run a two-week queue experiment

Begin with one service category whose jobs are reasonably comparable. Count open work at the same time each day and record completions daily. Keep cancelled jobs visible so they do not disappear from the story.

1. Write the process start and finish in one sentence.
2. Count daily open jobs and average them over the observation period.
3. Divide completed jobs by the number of working days in the same period.
4. Compare calculated elapsed time with actual start-to-finish dates for completed jobs.
5. Remove one recurring waiting reason and repeat the measurement.

Keep the improvement useful

Review customer complaints, rework and missed promises beside queue size. A smaller queue with more rejected work or rushed quality is not automatically a better business. Separate urgent repairs from standard repairs if they follow very different paths.

Keep the blockage, responsible person and next check beside each job. Confirm progress with the person doing the work, especially when a status changed during a call. Compare the queue at the same time each week so an apparent improvement is not simply a different counting method.

Save the reason for an exception while everyone still remembers it.